First Year Engineering Students’ Production &
Perception of Creativity in Design Ideas
Colin M. Gray1, Seda Yilmaz1, Shanna R. Daly2,
Colleen M. Seifert2, & Richard Gonzalez2
1 IOWA STATE UNIVERSITY
2 UNIVERSITY OF MICHIGAN
CREATIVITY & ENGINEERING
• Increase in design activities for first-year
engineering students (Dym & Little, 2004; Ogot & Okudan, 2006)
• Pedagogical tools and strategies have been
developed to increase creative potential
(Dym et al., 2005; Tolbert & Daly, 2013)
Specific barriers students face in learning how to be
creative are unclear (Csikszentmihalyi, 1988; Tolbert & Daly, 2013)
• Fixation as one diagnostic outcome of the lack of
creative ability (Purcell & Gero, 1996)
• Tools can be used to effectively discourage
fixation and increase capability
(Daly et al., 2010; Smith & Linsey, 2011; Smith, Linsey, & Kerne, 2011)
• Inaccurate beliefs about creativity slow the
development of creative ability
• Inaccurate beliefs have been studied in other
educational domains as “bugs” or “misrules”
(Brown & VanLehn, 1980; Engelmann, 1993)
• Such beliefs are deeply held, and difficult to document
or change without deep knowledge of how beliefs are
built and systemically altered (Carnine & Becker, 2010)
We are assuming that students’ tacit biases about
creativity shape their labeling of designs as “creative,” and
may serve as a barrier to their creation and recognition of
truly novel and useful concepts
1. Where do first-year engineering students
identify their “most creative” idea in an idea
generation activity, when compared with all
2. What characterizes the relationship between a
participant’s espoused belief about creativity
and their ordering of creative concepts?
3. Does Design Heuristics as a pedagogical tool
have an impact on the student’s perception of a
Provides prompts to help
alternatives that vary in
nature, discouraging fixation
and encouraging divergent
patterns of thinking
(Yilmaz, Daly, Seifert, & Gonzalez,
2011; Yilmaz, Seifert, & Gonzalez,
Derived from empirical
evidence of industrial and
(Daly et al., 2012; Yilmaz, Christian,
Daly, Seifert, & Gonzalez, 2012;
Yilmaz & Seifert, 2010)
Validated through a range of
product analysis, case
studies, and protocol
analyses, in both
(e.g., Yilmaz & Seifert, 2009; Yilmaz et
al., 2011; Yilmaz et al., 2010; Yilmaz
et al., 2013; Yilmaz, Daly, Christian,
Seifert, & Gonzalez, 2014)
• 156 first-year engineering students
• Two-day design-build-test activity prior to
matriculation at a large research university
• 85-minute idea generation session, using
brainstorming and Design Heuristics methods
• Survey on beliefs about creativity, ranking their
concepts from most to least creative
spill-proof coffee cup
spill-proof coffee cup
Generate as many
ideas as possible
Generate as many
additional ideas as possible
RELATIVE LOCATION OF
“MOST CREATIVE” CONCEPT
M (BS) M (DH)
I (BS) 21 (38.1%) -0.76 (0.05) 4.19 2.76
II (BS) 54 (70.4%) -0.30 (-0.27) 4.91 3.10
III (DH) 50 (60.0%) 0.37 (-0.32) 3.72 2.68
IV (DH) 31 (80.6%) 0.77 (-0.31) 4.55 3.06
TOTAL 156 (65.4%) 0.06 (-0.25) 4.36 2.91
EXAMPLE CASE: BROOKE (I)
A section of the roof of the car can be
brought down so it is in the back seat
or the floor when the seats are down.
The bikes can be attached to the roof
on clamps that are located on the roof
already. Once attached, controls lift
the roof back into place, with the bikes
MOST CREATIVE LEAST CREATIVE
EXAMPLE CASE: LINDA (IV)
MOST CREATIVE LEAST CREATIVE
The bike is fastened to a rack that is
just above ground level (attached with
locks like any normal bike). The
ground-level rack collapses to a rack
sticking off the back of the car, and
then the rack collapses in so the bike is
above the car.
Creative Concepts Must Never Have
Been Thought Of Before
Creative Concepts Must Be As Little
Like Shipping Products as Possible.
Creative Concepts Are
Creative Concepts Must Be
DESIGN HEURISTICS TO
Creative Concepts Can Be Based
on Existing Ideas
Creative Concepts Can Improve
on Shipping Products
Creative Concepts Can Be Practical
Ordinary Concepts Can Have
This research is funded by the National Science
Foundation, Division of Undergraduate Education,
Transforming Undergraduate Education in Science,
Technology, Engineering and Mathematics (TUES
Type II) Grants # 1323251 and #1322552.
First-year engineering students often fixate on ideas during the idea generation process, but it is unclear how they decide which of their concepts are most creative to pursue, and where their perceived “most creative” ideas occur in this process. These beliefs about creativity impact the ways in which students are able to conceive of and develop creative and novel solutions to the big engineering challenges of our age [1,2].
First year experiences for engineering students have been dramatically expanded in the past decade to include situated design activities, support for developing technical skill alongside mathematics and physics instruction, and team-based or collaborative design projects . Engineering students have cited project-based learning as important to their understanding of the engineering discipline, and a lack of contextual experiences has been linked to lower retention , underscoring the importance of student engagement in authentic engineering activities. To this end, many scholars have suggested pedagogical tools and strategies to engage students in these early experiences [5,6], including curricular structures  and specific strategies to increase creative potential [8,9]. It is less clear, however, which specific barriers beginning students face in learning how to be creative, especially in relation to existing beliefs they hold regarding the nature of creativity [6,10].
The most frequent framing for the discussion of creativity and creative development in engineering education is the barrier known as fixation—the inability to create concepts that diverge from known examples [13,14]. As with the methods for encouraging the development of creativity, many of these methods have been used more directly to discourage fixation , such as through the generation of more, and more varied, concepts with Design Heuristics , analogical thinking [17,18], or the decomposition of existing products .
But within all of these approaches, particularly in addressing fixation as one diagnostic outcome of a student lacking creative ability, we don’t have a clear understanding of what students believe, perhaps inaccurately, about creativity and how those beliefs might influence behavior. Adequate knowledge of these processes is critical to helping students become more creative.
These issues of misconception or misdirected generalization have been addressed in the broader educational community, often through the lens of mathematics [e.g., 25,26] or educational philosophy . Scholars in this area have highlighted how students come into the curriculum with “bugs”  or “misrules”  that they then apply in their work. These intrinsic beliefs are deeply held, and are difficult to change without direct instruction [26,28]. In particular, the misrules that students use, or perform, in learning activities are difficult to address apart from a systemic correction of the misrules through worked examples, similar to the student- or artifact-centered desk critique common in most design disciplines . We view misconceptions about creativity through much the same lens, assuming that students’ tacit biases about creativity shape their labeling of designs as “creative,” and may serve as a barrier to their creation and recognition of truly novel and useful concepts [29,30].
Our goal in this study was to explore student-identified “most creative” concepts by examining the temporal emergence of beliefs about creativity, production of design concepts, and idea generation methods. Within this framing, our analysis assesses the following research questions:
Participants were exposed to two idea generation techniques: brainstorming and Design Heuristics during an 85-minute session. Students were asked to individually generate concepts for one of two engineering problems: a bike rack and a spill-proof coffee cup . These two problems were chosen based on their previous use in the research literature to investigate creative idea generation in an engineering context. Students were also exposed to two different forms of idea fixation: a provided existing design solution to the problem given, and their own first ideas they generated prior to formal ideation. Each of the four experimental conditions, comprised of the two problem types and two forms of fixation, were carried out in separate sessions with a random subset of participants. In all subsets, students first used traditional brainstorming techniques in the first half of the session, generating as many ideas as possible. In the second half, the Design Heuristics method was introduced, and students were asked to use a randomly selected subset of the Design Heuristics cards to generate as many additional concepts as possible for the same problem. At the conclusion of the activity, students were asked to complete a short survey and rank all of the concepts they had generated through both methods from most to least creative, along with identifying generated concepts that were similar to the fixation source (either participant-created or provided). Participants then completed a survey about their beliefs on creativity.
Students were also exposed to two different forms of idea fixation: a provided existing design solution to the problem given, and their own first ideas they generated prior to formal ideation. Each of the four experimental conditions, comprised of the two problem types and two forms of fixation, were carried out in separate sessions with a random subset of participants.
In all subsets, students first used traditional brainstorming techniques in the first half of the session, generating as many ideas as possible. In the second half, the Design Heuristics method was introduced, and students were asked to use a randomly selected subset of the Design Heuristics cards to generate as many additional concepts as possible for the same problem.
Our initial analysis focused primarily on (1) the temporal location of the concept that each participant rated as most creative, (2) the method the participant used, and (3) the qualities of the most creative concept in relation to other concepts by that participant.
We then calculated the relative location of the identified “most creative” concept for each participant within the 85-minute idea generation activity (Figure 2). If the most creative concept was identified within the brainstorming (BS) phase, the variable ranged from -1 to 0; similarly, if the most creative concept was identified in the Design Heuristics (DH) phase, the variable ranged from 0 to +1. To further situate the location of the most creative concept without having tracked the actual time of creation, the variable range was divided by the total number of concepts each participant generated in that phase, and the midpoint of the most creative concept was calculated numerically. Without any data on the exact creation time for each concept, calculating an average allowed us to compare relative positions for further analysis. For instance, if a participant generated three concepts in the brainstorming phase and the middle concept was chosen as the most creative, the corresponding most creative concept variable would be calculated as -0.5.
Using this calculated variable, we then divided each method phase once more, to separate all participants that identified their most creative concept in the first or second half of each method, relative to the total number of concepts generated in that phase by each participant. This process yielded four subsets: I-BS (n=21), II-BS (n=54), III-DH (n=50), and IV-DH (n=31). Descriptive statistical analysis was performed on each subset (Table 1).
Summary of descriptive stats for each subset, which we used to derive exemplary cases.
The 156 participants included in our analysis comprised 102 male and 54 female students, 17 to 18 years of age. In total, these participants generated 1134 concepts (M=7.27; SD=2.69; min=2; max=15) across both idea generation methods. Participants in the four subsets were equally distributed across problem and fixation types.
Using these descriptive statistics (Table 1), we can explore several characteristics of participants who identified their “most creative concept” from various portions of the overall ideation exercise. Table 1 includes the number of participants in each subset, the mean location for the “most creative” and “least creative” concepts, and the mean number of concepts generated within each idea generation method. Participants in II generated more total concepts (M=8.01) than participants in IV (M=7.61), with participants in the other two phases generating fewer concepts (I: M=6.95; III: M=6.40).
Finally, we selected a corresponding representative case from each subset, with the resulting four cases including diversity of gender, total number of concepts generated, and location of the participant-identified most creative concept. Each of these cases will be presented in a later section, including further analysis of the progression of ideas in relation to the source of fixation, the nature of the relationship of the most creative concept to the fixation source, and stated beliefs regarding creativity from the exit survey in relation to the concepts generated.
Brooke is an 18-year-old female who generated concepts for the bike rack problem. In total, she generated eight concepts—four in each phase—after receiving an example bike rack solution as a form of priming fixation. Brooke labeled her most creative concept (Figure 4, left) as a moveable bike rack:
This bike rack represents an impractical idea—in that it modifies the car itself—but is seen by Brooke as a more creative option than her lowest ranked concept (Figure 4, right), a modification of the mounting portion of the rack: “The bike rack can be folded in half when it is not being used. The supports are folded down into ‘x's’. These supports are then used in multiple ways & different positions.” Within the overall ranking of concepts (from most creative to least creative), only one of the top five creative concepts out of the 8 concepts proposed was generated using the Design Heuristics method; solutions included the moveable roof (n=1), adjustable attachments for wheel mounting (n=2), a rack embedded into the top of the roof (n=1), and a multipurpose portion of the rack that could be used as a stepping stool (n=1). Only this final concept was created using Design Heuristics, with heuristic #10 (“allow user to reconfigure”). All three of the lowest ranked concepts were generated using Design Heuristics, with two of the three also indicated as being similar to the example fixation concept. Solutions included an indent in the car roof (n=1), a portion of the rack that can attach and pull a bike up from ground level (n=1), and the foldable rack (n=1).
Brooke described her approach to assessing creativity in her concepts as follows: “The concepts that I consider more creative are the ones that seem less practical/ realistic. They are harder to design, but much easier to use.” This belief about creativity is borne out in her highest ranked concepts. While the progressive improvements suggested in some lower ranked concepts might represent more creative pathways (e.g., using a portion of the rack as a step stool), the least realistic concept—requiring the most modification to the vehicle itself—was judged as “most creative.” Brooke’s conception of progressive enhancement seems to reinforce her beliefs about what constitutes a creative concept—where similar concepts are those where “I used the same type of design in a new way.” In her summary regarding learning about idea generation in the activity, she again linked unrealistic ideas with creativity: “I learned that unrealistic ideas can be used to generate a very realistic/creative product.”
The concepts Brooke created when using the Design Heuristics cards represented more incremental changes from the example fixation this group was provided; three of the four concepts generated with this method were marked as similar to the example concept—and presumably, based on the creativity ranking Brooke provided—less creative than the more divergent ideas generated in the brainstorming phase. The specific application of Design Heuristics was often unclear in Brooke’s concepts. While her use of #10 (“allow user to reconfigure”) was clearly evident in her repurposing of the rack as a stepping stool, other uses, such as #73 (“multiple components for one function”) for the lowest ranked foldable rack, provided less insight into how Brooke conceptualized the use of heuristics in relation to her concept.
Linda is an 18-year-old female who generated concepts for the bike rack problem. In total, she generated nine concepts—five in the brainstorming phase, and four in the Design Heuristics phase—after receiving an example bike rack solution as a form of priming fixation. The concept Linda identified as her most creative (Figure 9, left) was a rack that moved the bike from ground level to the top of the car:
This concept was created at the end of the Design Heuristics phase, using heuristic #65 (“telescope”) to define horizontal and vertical “collapsing” of the moveable rack in order to facilitate movement of the bike on the car. Linda’s lowest ranked concept (Figure 9, right) was a more traditional rack concept, using magnets to attach the rack to the vehicle and bicycle to the rack. Three of her top four concepts were generated using Design Heuristics, including the top-ranked moveable rack (1), a rectangular prism used to hold the bike into place (1; using #66—“texturize”), and fastening the bike on its side rather than standing up (1; using #75—“use recycled or recyclable materials”),). Her bottom three ranked concepts, all of which were generated during the brainstorming phase, included ways to move the bike into place automatically (2) and the use of magnets to attach the rack (1). The center two concepts included another iteration of the moveable rack generated using brainstorming, and a collapsible roof area to contain the bike, generated using Design Heuristics #65 (“telescope”).
Linda enunciated the broadest definition of creativity of these four cases, asserting:
My most creative concepts came from the heuristics cards. They allowed me to think outside of the box and come up with new ideas. I also think that not analyzing while brainstorming helped me significantly. It allowed me to get my ideas down before comparing it to others.
She noted that she had seen bike racks attached to the top and back of cars before, and this may explain two of her moving rack concepts from the brainstorming phase where the rack originates at the back of the vehicle, both of which she marked as being similar to the priming example. Linda directly cites the Design Heuristics cards as an aid in more divergent ideation, noting that the brainstorming concepts “are very similar because I did not have the heuristic cards (so I did not think as much out of the box).”
Linda appears to make use of Design Heuristics in refining earlier concepts that were generated in the brainstorming phase—making the specific function that she had initially documented clearer or better designed. For instance, the telescoping motion Linda describes in her last and highest ranked concept, where the rack collapses in a vertical, then horizontal direction, was present in her very first concept of the brainstorming phase (which was ranked eighth). The highest ranked concept includes a specific way for the rack to change positions, while in the lower ranked, only a general movement along a track is described. Interestingly, Linda notes the similarity of two other concepts—both from the brainstorming phase (Figure 10)—that include some movement of the rack from the side or back of the car to the top; but even though this general concept of movement is shared with the top-rated concept, no similarity is noted in that concept. As such, Linda appears to demonstrate some preference for concepts generated later, even where the core idea was articulated earlier, albeit with less specificity.
Creative Concepts Must Never Have Been Thought Of Before. This misrule is best demonstrated by Clark, who assessed creativity based on whether “ive [sic] seen it before.” This belief about creativity often leads to a weak chain of iteration between concepts, with a particular struggle in generating an initial concept that is judged sufficiently “creative” vis-à-vis existing concepts on which to begin generating alternatives . In practice, as demonstrated by Brooke, this misrule can lead to outlier concepts that disregard identified constraints (e.g., moveable roof to attach and lift bike), which are difficult to assess or iterate from.
2. Creative Concepts Must Be As Little Like Shipping Products as Possible. Starting with this misrule, participants judge existing solutions—no matter how novel—as uncreative, although often tacitly so. Instead of being able to identify components or approaches of existing products that may be modified to create an equally creative concept, participants attempt to create concepts that are extremely different from existing products on the market (e.g., Andy’s gourd-shaped vessel with a cork).
3. Creative Concepts Are Generally Impractical. Embedded within this misrule is the latent assumption that existing concepts are practical concepts—in other words, they are technologically or economically possible. So therefore, concepts which are not currently possible, due to available materials and manufacturing processes, or other factors, must inherently be creative.
4. Creative Concepts Must Be Completely Creative. This misrule is perhaps the most nuanced, but can have a dramatic effect on the quality of divergence that students are able to create in their concepts. Similar to the first two misrules, this one starts with the tacit assumption that the product must not have been thought of before, or be totally different than existing products. Within this perspective, embodied by all four cases to some degree, the holistic product is being assessed—and its match to other holistic products. This leaves out the opportunity for progressive enhancement through the creative addition of a component or product function, and results in potentially creative solutions that bridge off of existing products being seen as not very creative, at least in comparison to “completely new” concepts.
For example, even in the case of Linda—an otherwise exemplary case—we see a pathway of concepts that is not entirely realized; while creative ideas emerged very early in the idea generation process, she did not recognize this idea as creative (as evidenced by her ranking) until the technical issues surrounding specific functional concerns were addressed, often one or two iterations after the core idea was created.
While many of these misrules are deeply embedded in students’ approach to generating and assessing ideas, Design Heuristics has been successfully used to promote the generation of divergent concepts in engineering contexts, producing a greater quantity of concepts which are of higher quality than other concepts produced through brainstorming alone. Because the focus of Design Heuristics is the application of heuristics to generate early design solutions, which function as modifiers to existing design concepts, it is an ideal tool to communicate the nature of idea pathways, and the relationship of these pathways to the misrules discussed above. We will briefly discuss how Design Heuristics may be used—not only as an idea generation tool, but also as a guided instructional tool to directly combat misrules about creativity.
Creative Concepts Can Be Based On Existing Ideas. To demonstrate that iterating from existing concepts can generate creative concepts, Design Heuristics can be used to progressively transform an existing concept into a revolutionary one by applying a different heuristic in each stage to alter some component or portion of the concept . By comparing the starting and ending concepts, the student can then articulate the idea pathways that were followed, demonstrating that even a series of small changes can result in the complete transformation of a concept .
Creative Concepts Can Improve on Shipping Products. Through progressive enhancement or component redesign, Design Heuristics can be used to identify and reimagine components of an existing, successful product, resulting in a related product that is creative in its own right. Decomposition techniques such as functional decomposition or morphological analysis can be used to identify a range of components within a larger design. Through the selection of a specific component, such as an existing “pain point,” a scoped idea generation activity can be framed, allowing the student to reimagine that component in isolation, encouraging the development of alternatives to an existing approaches.
Creative Concepts Can Be Practical. An approach to address this misrule might be to identify what is truly innovative about the concept. Qualities of materiality and functionality can both be countered through targeted ideation, either through adaptation of extant morphological analysis approaches, or through adoption of a subset of Design Heuristics cards. The sci-fi or futuristic concept can also be beneficial, but must be redirected into implications for concepts that are actually possible to create; for instance, what social or experiential qualities of a given “fantasy” device might be desirable, and how those qualities might map (in preliminary form) to products that can be designed [36,37].
Ordinary Concepts Can Have Creative Components. Similar to the improvement of existing products, seemingly ordinary products can contain highly creative components. Decomposition techniques may be used to identify such components in existing products, which can then be applied in new and unexpected contexts. Design Heuristics can also be a useful tool in combining ideas or components from multiple products, including strategies for synthesizing functions or recognizing and altering the use qualities of a product based on external or contextually-dependent characteristics. Students appear to confuse “novel” as the sole criterion for creative, while “useful” or “having value” is another  As a result, they may reject concepts with familiar components even though highly novel components are included.